Comparative performance analysis of ChatGPT 3.5, ChatGPT 4.0 and Bard in answering common patient questions on melanoma

This study evaluates the effectiveness of ChatGPT versions 3.5 and 4.0, and Google's Bard in answering patient questions on melanoma. Results show that both versions of ChatGPT outperform Bard, particularly in readability, with no significant difference between the two ChatGPT versions. The study underscores the potential of large language models in healthcare, highlighting the need for professional oversight and further research.